Partially Verified65% confidenceBenchmarkExact time
OpenAI 称 GPT-5.6 在 Terminal Bench 2.1 上达到当前最强水平
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
7/10/2026
First Seen
Valid until: 10/8/2026
Sources
GPT-5.6三模型实测：Sol/Terra/Luna与Claude差距还有多远？
bilibili烟神殿副殿主7/9/2026
Related Claims
UnverifiedGPT 5.5 leads on TerminalBench 2.0 with a score of 87%, surpassing even Anthropic's unreleased Mythos model80% similarUnverified6月26日OpenAI发布GPT-5.6系列，首次采用旗舰、平衡、快速、廉价四档分层策略，旗舰版本在Terminal Bench 2.1基准测试中以91.9%成绩登顶80% similarUnverifiedOpenAI称GPT-5.6 Sol在Terminal-Bench 2.0上达到了新水平79% similarUnverifiedOpenAI在GPT-5.6发布材料中回避了SWE Bench Pro的分数78% similarUnverified在OpenAI官方公布的基准测试中，GPT-5.6在编程领域Terminal Bench拿下新的最高分，生物学基准测试比GPT-5.5更省Token（成绩为OpenAI自行披露，尚未独立验证）77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/452863API
curl https://kongchang.com/api/v1/knowledge/claims/452863MCP
get_claim(id=452863)